Have you noticed the hyper euphoria related to the move towards a ‘cloud’ based environment? Have you questioned if this strategy is all that it is said to be? You really should.
If the cloud is so great, why are so many businesses unsatisfied?
Cloud computing has become common in enterprise IT, and the hype around it remains as adoption soars. Research by IDG shows that 70 percent of enterprises currently use at least one cloud application, and in 2018, organizations with cloud-only IT infrastructure will become the majority.
The global market for cloud services was worth $148 billion in 2016, according to Synergy Research Group, and it is growing by 25 percent annually. Amazon Web Services (AWS) alone reached $3.23 billion in revenue in the third quarter of 2016, while Microsoft Azure, the second-largest cloud provider, announced Thursday that revenue has nearly doubled in the past year, giving it an annual run rate of $14 billion.
Over a third of enterprises find that cloud implementations fail to live up to the hype, and half say they deliver only some of their expected benefits, according to a survey of CIOs by Vanson Bourne. What is the reason for the disappointing results?
According to more than three out of four (77 percent), the difficulty is in finding and deploying a service that is the right fit.
Failure to find and deploy the right environment contributes to costly unused cloud resources, which are experienced by 39 percent of enterprises, according to a 2015 report by Forrester.
Further, a survey by Enterprise Management Associates found that unsuccessful cloud implementations with major cloud providers are common, with over half of attempts with AWS stalling or failing.
For many organizations, a ready-made premium cloud environment is a better fit than hyper-scale providers that offer a huge range of tools for organizations to use in crafting their own.
So, what are the issues being faced to adopt a multi-cloud environment?
IDC predicted that more than 85 percent of enterprise IT organizations will adopt multi-cloud architectures by 2018.
But for all the benefits of a multi-cloud strategy, there are some challenges that come with it as well. Specifically it can be difficult to secure a multi-cloud strategy because of a lack of visibility across hosts and services. That makes it easier for hackers to find exploitable vulnerabilities within an organization’s infrastructure and it also makes it more difficult to meet compliance mandates.
A cloud service has clear advantages, no argument. But, there can be several serious technical challenges too.
When you use Software as a Service (SaaS) from multiple vendors, it can get tricky. Most of the time, for users in North America, it works relatively well. But, sometimes, there can be issues of latency, data rates, and time-outs.
If you have one application such as an Enterprise Resource Planning (ERP) solution hosted in Northern California, and a second solution, perhaps a Human Resource Management (HRM) solution hosted in Miami, and a third solution such as an Asset Management (AM) solution hosted in Toronto, the issue may be the delays between the requisite inter-cloud APIs to connect between applications. If these applications resided on a single cloud, the issues are largely mitigated. The same can be said in the past when they were hosted in your private data center. You need an Application Protocol Interface (API) to permit the exchange of data in real-time between these applications. The distances, security, data conversion, and data rates may cause issues for you if these connections are not designed perfectly. A third-party API management solution is a smart option. It configures, manages, and grooms the API traffic between disparate cloud points of presence (POPs).
By rationalizing the cloud sites and bringing multiple clouds to a common geographic area, you can reduce latency, address legislation jurisdictions, and provide for better inter-cloud coupling.
Another issue is how you connect to these disparate cloud hosting sites. Most allow a dedicated software defined wide area network connection or an IP/MPLS pipe to directly tie your company to the cloud location. However, what if these hosted application will not accept a dedicated connection. Some cloud providers only permit a connection to the public internet with a VPN. So, you may need to bring your applications together with different types of connections which means different levels of performance. These performance differences can cause havoc with different latency performance between connection types.
A third concern is how you implement your security. In the old days, we collocated all the applications inside a shared data center and then build a multi-million dollar firewall around this data center to protect everything within the facility. As an analogy, it was like building a moat around the castle. Once you crossed inside over the draw bridge you had untethered access to everything within the castle. If the proverbial Trojan Horse got over the bridge, you were in big trouble from an attack within your fortress.
However, now with a ‘Zero Trust’ model we have inverted the security paradigm and trust absolutely no one. We have gatekeepers everywhere within these myriad of data centers to validate user rights, permit access, and log all movements to a centralized historian. So, in this case, the Zero Trust model is said to be better. However, in a multi-cloud environment, these gatekeepers must be federated over this disparate cloud model too. You need to create gatekeepers on each cloud POP and then centralize the data back to the core centralized scrutineer for analysis and approval. This adds to latency and can make connections untenable.
Governmental regulations may or will likely handcuff what you can do and where you can do it. Hosting data at different cloud location in different geopolitical jurisdictions might demand adherence to different legislation and complex rules for privacy. For example, with the new GDPR privacy rules in the European Union, which are not aligned to the regulations in the USA or India or Asia, you may be facing some serious challenges for legislative compliance.
Compliance standards are of utmost importance, irrespective of the type of Cloud services being utilized. When different Cloud providers are being employed, it is important to note if the various compliance standards like HIPAA, PCI DSS, FISMA, and SOX, are being met with each cloud vendor. If these compliance standards are not met, the data would be at risk of being hacked or lost. As well as a potential for fines and public embarrassment for the company. No company wants to be the reason for a harsh new headline.
Multi-cloud challenges might seem to be a lot; however, the benefits which can be obtained from multi-cloud environments can take an organization places. What is important is the proper implementation of these environments and the migration of applications to the private and public clouds which should be performed with utmost caution — the benefits will eventually supersede the challenges.
Beyond legislation issues, you may face conflict with your own corporate data governance policies in a multi-cloud setting.
The biggest challenge is to understand where the data resides physically – this situation might be graver for small and mid-size companies.
Given the multi-cloud environment functionality, it might be straightforward to make a mistake and end up running an application in an unapproved environment. There are loads of guidelines laid down, especially under GDPR, which when breached, can cause a collection of hefty fines; to curb this issue, IT managers might be required to prepare the right tools to garner visibility to monitor the extent of their regulatory burdens. Lack of situational awareness and poor data visibility make data governance onerous to manage. So, one of the key advantages of these ubiquitous clouds may also be one of the hardest issues to administrate.
Not all cloud operators are equal. Some offer excellent customer service capabilities and others lack abilities. Regardless, they are all different from one another.
When an organization is utilizing multiple Cloud vendors, there are a lot of factors which need to be considered. When multiple vendors are being used, multiple skill sets will also need to be managed to get the maximum out of the Cloud. The higher the number of SaaS, IaaS, and PaaS solutions leveraged, the higher would be the in-house skill sets which will need to be involved. Many customers who I speak to about multi-cloud environments look to reduce the internal organizational staffing burden and not increase it. But, a multi-cloud setting is a complex environment and your responsibilities will not be shifted away, in fact they may actually increase for your business.
Even though multiple vendors are a necessity for managing the workload within large IT teams, it is important to strive for commonality, wherever possible. Such a feature can be exercised by making use of conventional operating systems, system administration tools, and development languages to manage Cloud management platforms.
In response to all these challenges, enterprises frequently start shopping for a hybrid or multi-cloud monitoring and management tool. But these tools can be expensive. In addition, they offer such different and wide-ranging feature sets that comparisons can be difficult. For example, Gartner’s Evaluation Criteria for Cloud Management Platforms and Tools identifies 215 different criteria that IT decision makers need to consider when selecting multi-cloud monitoring tools. That level of research and evaluation takes a significant level of effort.
Most organizations already have monitoring tools that they use for their in-house data centers. In a perfect world, they might want to extend their use of these tools into the cloud.
However, most legacy tools were never designed for cloud monitoring. While some legacy vendors have updated their monitoring tools to provide some cloud support, most fall short of providing the level of detail and comprehensive capabilities that many enterprises are seeking.
Multi-cloud management can be very hard to do well, maybe even impossible to do gracefully. A unified approach and a common administration is desired. But can it be achieved? You may be faced with several disparate service management tools or a hierarchical ‘Manager of Manager’ stacked set of tools to manage a complex multi-cloud setting. Again, a complicated and difficult to manage situation.
Every vendor deployment comes with its own unique portals and processes that companies need to manage. Consider that even something as seemingly simple as Identity and Access Management can be complicated if different providers demand differing password complexities or authentication measures.
The simplest solution here? A multi-cloud management platform that brings unique resources under a single umbrella, and helps avoid problems with platform and process sprawl. While a good start, such platforms typically are never a one-stop shop. Meaning, you might find great multi-cloud monitoring platform, but still rely on manual security patching or provider portals for advanced configuration.
Multi-cloud environments make it easier than ever to lose track of which applications are running, where and how much this costs you day to day. For example, employees may be partial to a particular cloud for certain workloads, but across your entire staff, this choice may not be consistent. This means you could end up with three, four, or more of the same app, open across multiple clouds.
Mastering multi-cloud economics is perhaps the greatest challenge of all. Each platform has its own unique set of variables that make optimization a full-time job – billing systems, pricing models, instance/VM sizing differences, data egress fees, etc. The multi-vendor sprawl of billing itself can induce budgeting nightmares for IT management. Partnering with a managed multi-cloud provider to consolidate billing and provide application-specific cost analysis, however, can simultaneously ease a lot of this burden and keep IT’s relationship with the finance department healthy. (Always a good idea!)
References:
Harvey, C. (2018). Top 5 Challenges of Monitoring Multi-cloud Environments. UBM Americas, a UBM plc company. Retrieved on August 2, 2019 from, https://www.networkcomputing.com/cloud-infrastructure/top-5-challenges-monitoring-multi-cloud-environments
Idexcel Industries. (2019). The Challenges of Multi-Cloud Environments. Idexcel, Inc. Retrieved on August 2, 2019 from, https://www.idexcel.com/blog/the-challenges-of-multi-cloud-environments/
ThinkIT. (2017). The Top Challenges Facing Multi-Cloud Environments. Internap Corporation. Retrieved on August 2, 2019 from, https://www.inap.com/blog/top-challenges-multi-cloud-environments/
Tkatchuk, R. (2017). If the cloud is so great, why are so many businesses unsatisfied? CIO Magazine, IDG Communications. Retrieved on August 2, 2019 from, https://www.cio.com/article/3163967/if-the-cloud-is-so-great-why-are-so-many-businesses-unsatisfied.html
About the Author:
Michael Martin has more than 35 years of experience in systems design for broadband networks, optical fibre, wireless and digital communications technologies.
He is a Senior Executive with IBM Canada’s Office of the CTO, Global Services. Over the past 14 years with IBM, he has worked in the GBS Global Center of Competency for Energy and Utilities and the GTS Global Center of Excellence for Energy and Utilities. He was previously a founding partner and President of MICAN Communications and before that was President of Comlink Systems Limited and Ensat Broadcast Services, Inc., both divisions of Cygnal Technologies Corporation (CYN: TSX).
Martin currently serves on the Board of Directors for TeraGo Inc (TGO: TSX) and previously served on the Board of Directors for Avante Logixx Inc. (XX: TSX.V).
He serves as a Member, SCC ISO-IEC JTC 1/SC-41 – Internet of Things and related technologies, ISO – International Organization for Standardization, and as a member of the NIST SP 500-325 Fog Computing Conceptual Model, National Institute of Standards and Technology.
He served on the Board of Governors of the University of Ontario Institute of Technology (UOIT) [now Ontario Tech University] and on the Board of Advisers of five different Colleges in Ontario. For 16 years he served on the Board of the Society of Motion Picture and Television Engineers (SMPTE), Toronto Section.
He holds three master’s degrees, in business (MBA), communication (MA), and education (MEd). As well, he has diplomas and certifications in business, computer programming, internetworking, project management, media, photography, and communication technology.